152 IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, VOL. 58, NO.

1, JANUARY 2009
Detection of Flicker Caused by Interharmonics
Taekhyun Kim, Student Member, IEEE, Edward J. Powers, Fellow, IEEE, W. Mack Grady, Fellow, IEEE, and
Ari Arapostathis, Fellow, IEEE
AbstractInternational standard bodies have specied an in-
strument capable of detecting and assessing the severity of lighting
icker, which is called a ickermeter. However, a deciency with
regard to icker caused by high-frequency interharmonics has
been reported in the recent literature. In this paper, another
deciency of the International Electrotechnical Commission (IEC)
ickermeter with regard to low-frequency interharmonics, whose
frequencies are below the difference between the fundamental
frequency and the cutoff frequency of a bandpass lter of the
IEC standard ickermeter, will be discussed and illustrated based
on analysis and numerical experiments. A new approach based
on down-up sampling in the discrete-time sample domain is also
proposed to address the icker-detection problem associated with
interharmonics. We demonstrate the efcacy of the proposed
method by a comparison with the current ickermeter standard
in the presence of interharmonics.
Index TermsDown-up sampling, icker, ickermeter,
harmonics, interharmonics, peak detection.
I. INTRODUCTION
I
N ELECTRIC power distribution networks, various types of
disturbances may occur, and they can affect electric power
customers in different ways. Among those disturbances, light
icker might be the most visible one for customers. Light
icker is dened as noticeable illumination changes due to volt-
age uctuations imposed on the fundamental power frequency
component. In addition to the fact that the illumination level
changes may signicantly impact work efciency, they might
also induce seizures in certain people [1]. Therefore, health
and safety issues with regard to icker have been raised. Fur-
thermore, the voltage uctuations associated with icker can
also cause potentially serious problems to critical power system
components such as motors, generators, and transformers.
Therefore, icker detection and evaluation has garnered the
attention of international standardization bodies. The Interna-
tional Electrotechnical Commission (IEC) specied an instru-
ment capable of measuring the severity of icker, which is
called a ickermeter [2], and the IEEE recommended the adop-
tion of the IEC ickermeter standard [3]. The IEC standard-
compliant ickermeter simulates human eyebrain responses to
Manuscript received July 7, 2007; revised May 13, 2008. First published
August 19, 2008; current version published December 9, 2008. This work
was supported by the U.S. Ofce of Naval Research under Grant N00014-02-
1-0623. This paper was presented in part at the 2007 IEEE Instrumentation
and Measurement Technology Conference, Warsaw, Poland, May 13, 2007.
The Associate Editor coordinating the review process for this paper was
Dr. Richard Thorn.
The authors are with the Department of Electrical and Computer Engineer-
ing, The University of Texas at Austin, Austin, TX 78712-0240 USA.
Color versions of one or more of the gures in this paper are available online
at http://ieeexplore.ieee.org.
Digital Object Identier 10.1109/TIM.2008.928413
illumination ux changes of lament (i.e., incandescent) lamps
due to voltage uctuations.
Flicker is usually produced by large-capacity, time-varying
loads such as adjustable-speed drives (ASDs) and arc furnaces.
These loads often generate noninteger harmonics of the fun-
damental power frequency, i.e., interharmonics, and it is well
known that icker phenomena are closely related to those
interharmonics [4][6].
However, according to recent studies [4], [7], [8], the current
IEC standard-based ickermeter cannot detect icker caused
by interharmonics whose frequencies are higher than a certain
frequency, i.e., 102 Hz for 60-Hz systems and 85 Hz for
50-Hz systems. A real eld case documenting the deciency of
the current standard is reported in [8]. Utility customers com-
plained about light icker, but standard-compliant ickermeters
could not detect it. To address the ickermeter standards de-
ciency, several approaches have previously been proposed [4],
[9], [10], but they are based on empirical weighting functions
or deductive expansion from the current ickermeter standards.
In this paper, another deciency of the current IEC icker-
meter will be presented. Our analysis and numerical experi-
ments indicate that, in the presence of interharmonics whose
frequencies are below the difference between the fundamental
frequency and the cutoff frequency of a bandpass lter of the
IEC standard ickermeter, the demodulated beat frequencies
detected by the IEC ickermeter do not match the actual beat
frequencies associated with the interharmonics. Various types
of frequency converters, including cycloconverters and ASDs,
can generate these low-frequency interharmonics [11], [12].
To address the icker-detection problem, we propose a novel
approach based on down-up sampling of the voltage signal.
Our proposed method detects peak uctuations imposed on
the absolute values of the voltage signal, and our analysis
and simulation experiments conrm that the proposed method
is able to detect icker, regardless of the frequency of the
interharmonics.
The remainder of this paper is organized as follows: The
current IEC ickermeter is reviewed in Section II. Deciencies
in the ickermeter response to interharmonics are discussed in
Section III. To overcome these deciencies in Section IV, we
propose a novel icker-detection method based on down-up
sampling, and the efcacy of the proposed method is veried
in Section V. Section VI concludes this paper.
II. IEC FLICKERMETER REVIEW
The IEC ickermeter estimates the severity of icker using a
voltage input signal. The icker estimation process is based on
the incandescent lamp model and the lampeyebrain response
0018-9456/$25.00 2008 IEEE
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE UBERLANDIA. Downloaded on May 11,2010 at 00:31:57 UTC from IEEE Xplore. Restrictions apply.
KIM et al.: DETECTION OF FLICKER CAUSED BY INTERHARMONICS 153
Fig. 1. Simplied diagram of the IEC standard-based icker estimation blocks [2].
model. Fig. 1 illustrates a simplied block diagram of the IEC
standard-based ickermeter [2].
Block 1 rst normalizes voltage input using a transformer
and a gain controller. The normalized voltage input signal is
squared by a multiplier in Block 2 to demodulate the icker-
related voltage uctuation term.
Block 3 contains two lters. The rst lter is a bandpass
lter for extracting the modulating terms from the squared
voltage signal. The bandpass lter consists of a rst-order high-
pass lter whose cutoff frequency is 0.05 Hz and a sixth-
order low-pass lter whose cutoff frequency is set to 42 for
60-Hz systems and to 35 Hz for 50-Hz systems. This cutoff
frequency is determined based on the assumption that any
frequency higher than the cutoff frequency does not cause a
signicant illumination change due to the thermal inertia of the
laments in incandescent lamps. The second lter in Block 3
is a weighting lter representing the human visual response
to uctuations. It is known that the human visual sensitivity
to illumination changes reaches a maximum level when the
frequency is between 8 and 10 Hz. Therefore, the lter response
in the second weighting lter of Block 3 has a maximum
spectral response at 8.8 Hz.
The signal ltered by Block 3 continues to be processed
by Block 4, which emulates human lampeyebrain responses
to illumination ux uctuations. The nonlinear and memory
effects of the human responses are simulated by squaring and
smoothing lters. The output of Block 4 indicates the instan-
taneous icker level. One unit of the output corresponds to
icker perceptibility threshold and is called the perceptibility
unit (PU).
The last block of the IEC ickermeter is the statistical evalu-
ator of the instantaneous icker level. Based on the cumulative
distribution function of the instantaneous icker level (PU), the
icker sensation is evaluated over a longer period of time. One
output of Block 5 is the short-term icker severity index P
st
.
P
st
is a measure of severity based on an observation over 10 min
[2]. One unit of P
st
corresponds to the threshold of illumination
changes causing irritation to people, which is the so-called
irritation threshold.
III. FLICKER CAUSED BY INTERHARMONICS
One potential limitation of the IEC ickermeter is its lack
of consideration for icker caused by interharmonics. It has re-
cently been shown that the icker characteristics and generation
process are considerably different in the presence of interhar-
monics [4]. As mentioned earlier, the current IEC ickermeter
is not capable of detecting icker caused by high-frequency
interharmonics [4], [7], [8]. According to [7], the deciency is
due to the squaring and ltering processes (in Blocks 2 and 3 in
Fig. 1, respectively) during the icker-estimation procedure of
the IEC ickermeter.
In addition to these deciencies, in this paper, we will present
another deciency of the IEC ickermeter with regard to icker
caused by low-frequency interharmonics, whose frequencies
are below the difference between the fundamental frequency
and the cutoff frequency of a bandpass lter of the IEC
standard ickermeter. With the low-frequency interharmonics,
the modulation (or beat) frequencies represented by the IEC
ickermeter are not consistent with the actual beat frequency.
We will demonstrate the discrepancy with simple analysis and
examples in this section.
The icker-evaluation process in the presence of interhar-
monics can be examined using a signal model with a funda-
mental frequency term and an interharmonic term, i.e.,
v(t) = sin(2f
1
t) + msin(2f
IH
t +
IH
), (1)
where f
1
, m, f
IH
, and
IH
are the fundamental frequency, in-
terharmonic relative magnitude, interharmonic frequency, and
phase of the interharmonic, respectively. We assume that the
input signal is already normalized. This interharmonic signal
model is also used in [6] and [7]. Note that the signal model in
(1) is different from the direct amplitude modulation models
(sinusoidal and rectangular voltage uctuations) used in the
IEC standard [2].
According to analytical and experimental studies in [4], the
actual beat frequency f
B
due to the interharmonic is known as
follows:
f
B
= |f
IH
hf
1
| = |f
IH
(2m + 1)f
1
| , m = 0, 1, 2, . . .
(2)
where h indicates the closest odd-order harmonic number
to f
IH
.
In [7], it is shown that if f
IH
is higher than 102 Hz for 60-Hz
systems or 85 Hz for 50-Hz systems, icker associated with the
interharmonic is not detected by the IEC standard ickermeter.
In this paper, we will focus on the case when f
IH
is lower than
102 or 85 Hz.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE UBERLANDIA. Downloaded on May 11,2010 at 00:31:57 UTC from IEEE Xplore. Restrictions apply.
154 IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, VOL. 58, NO. 1, JANUARY 2009
Fig. 2. Decient icker-detection example of the current IEC standard. (a) Waveform with an 8-Hz interharmonic. (b) DFT magnitude of the Block 3 output
of the IEC ickermeter. (c) Half-cycle RMS value uctuations. (d) DFT magnitude of the RMS uctuations. (e) Absolute value peak value uctuations. (f) DFT
magnitude of the peak uctuations.
According to the IEC ickermeter standard, the input voltage
signal is rst squared and then input to the bandpass lter in
Block 3. The squared input voltage signal in (1) becomes
v
2
(t) = [sin(2f
1
t) + msin(2f
IH
t +
IH
)]
2
=
1
2
(1 + m
2
)
1
2
cos(2 2f
1
t)

1
2
m
2
cos(2 2f
IH
t + 2
IH
)
+ mcos (2 (f
1
f
IH
) t
IH
)
mcos (2 (f
1
+ f
IH
) t +
IH
) . (3)
As a result of the bandpass lter in Block 3, the dc term and
frequencies higher than the cutoff frequency (42 or 35 Hz) in
(3) are ltered out. Therefore, only two frequency terms, i.e.,
2f
IH
and f
1
f
IH
, can possibly remain after ltering.
To simplify the discussion, we will only consider 60-Hz
systems (f
1
= 60) and the corresponding cutoff frequency
(42 Hz) of the bandpass lter in Block 3. We also assume that
the lters in the icker processing blocks are ideal. Depending
on the f
IH
value and the corresponding response of the IEC
ickermeter, the entire frequency range can be divided into
three regions, and we will look into the response of the IEC
ickermeter in each frequency region.
A. 0 < f
IH
18
In this case, 0 < 2f
IH
36, and f
1
f
IH
42. Hence,
only the 2f
IH
term remains after bandpass ltering. Thus,
the 2f
IH
term represents the uctuating term detected by
the ickermeter. However, the actual beat frequency associ-
ated with this interharmonic frequency f
IH
is not 2f
IH
but
f
1
f
IH
, according to (2), since the closest odd-order har-
monic number h is 1.
This discrepancy in the actual beat frequency and that de-
tected by the ickermeter can lead to inaccurate icker eval-
uation results. For example, the weighting lter following the
bandpass lter in Block 3 simulates the frequency response
of lampeye reaction to light icker. Therefore, if the input
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE UBERLANDIA. Downloaded on May 11,2010 at 00:31:57 UTC from IEEE Xplore. Restrictions apply.
KIM et al.: DETECTION OF FLICKER CAUSED BY INTERHARMONICS 155
Fig. 3. Compliance test result of the numerically implemented IEC ickermeter. Plot of the percentage deviation versus modulation frequencies.
frequency to the weighting lter is not matched with the actual
beat frequency, the resulting icker-evaluation indexes can
become erroneous with respect to actual icker phenomena.
Fig. 2 illustrates the beat frequency discrepancy of the IEC
ickermeter in the presence of a low-frequency interharmonic.
Fig. 2(a) and (b) presents a sample voltage input (f
1
= 60 Hz,
m = 0.2, and f
IH
= 9 Hz) and the discrete Fourier transform
(DFT) magnitude of the Block 3 output of the IEC ickermeter,
respectively. As discussed earlier, only the 16(= 2 8) Hz
term remains after bandpass ltering in Block 3 of the IEC
ickermeter. Fig. 2(c) and (d) illustrates the half-cycle root-
mean-square (RMS) uctuations and the corresponding DFT
magnitude, and Fig. 2(e) and (f) demonstrates the absolute
value peak uctuations and the corresponding DFT magnitude.
Since the input ac voltage is squared (e.g., incandescent lamps)
or rectied (e.g., uorescent lamps) to produce the light out-
put, the uctuations of the half-cycle RMS values and of the
absolute values should be considered. For both cases, the actual
uctuation terms are 52(= 60 8) Hz, as shown in Fig. 2(d)
and (f), which is not consistent with the DFT magnitude result
of the IEC ickermeter shown in Fig. 2(b). Furthermore, the
magnitude of the 16-Hz term in Fig. 2(b) is determined by the
square of the interharmonic relative magnitude, namely, m
2
,
which is very small since usually m 1.
B. 18 < f
IH
< 102
This frequency range can again be divided into two cases, as
follows.
1) 18 < f
IH
21: Both the 2f
IH
and f
1
f
IH
frequency
terms are left after ltering. However, the 2f
IH
term can be
ignored since the corresponding magnitude m
2
is very small,
as stated earlier. Therefore, the f
1
f
IH
term represents the
beating term.
2) 21 < f
IH
< 102: In this frequency range, only the f
1

f
IH
term in (3) remains after bandpass ltering. This frequency
is consistent with the actual beat frequency, as indicated in
(2), with h = 1. Thus, the IEC ickermeter appears to prop-
erly assess icker caused by interharmonics in this frequency
region.
C. 102 f
IH
The deciency of the IEC ickermeter in this frequency
range has been discussed by other authors [4], [7], [8], [13].
Fig. 4. Minimum interharmonic magnitude m
min
necessary to produce
one PU of the IEC ickermeter response versus interharmonic frequencies
ranging from 0 to 120 Hz.
Since the bandpass lter in Block 3 suppresses all interhar-
monic terms whose frequencies are higher than 102 Hz (85 Hz
for 50-Hz systems), the IEC ickermeter cannot detect icker
caused by interharmonics in this frequency range, regardless of
their magnitude.
To investigate the response of the IEC ickermeter to in-
terharmonics, we numerically implemented the IEC standard
ickermeter algorithm using MATLAB and experimented with
interharmonics ranging from 0 to 120 Hz. To ensure the com-
pliance of the implemented ickermeter to the IEC standard,
the compliance test based on the response of the ickermeter to
sinusoidal amplitude modulation was carried out according to
the Normalized ickermeter response for sinusoidal voltage
uctuations table in the standard [2], and the results are
presented in Fig. 3. Fig. 3 is a plot of the deviation of the
minimumvoltage uctuation level, which generates a unit value
of PU, from the reference values in the IEC standard table
versus different modulation frequencies. The IEC standard
allows a maximum 5% deviation, and the test results in Fig. 3
show that our numerical implementation is well within the 5%
allowance.
Fig. 4 illustrates the sensitivity of the IEC ickermeter to
interharmonics ranging from 0 to 120 Hz. Each point on
the curve in Fig. 4 represents the minimum interharmonic
magnitude m
min
, which results in a unit value of PU at the
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE UBERLANDIA. Downloaded on May 11,2010 at 00:31:57 UTC from IEEE Xplore. Restrictions apply.
156 IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, VOL. 58, NO. 1, JANUARY 2009
Fig. 5. Proposed icker peak detection blocks.
IEC ickermeter output at each interharmonic frequency. The
frequency range 0120 Hz is divided into three regions based on
the characteristics of the IEC ickermeter, as discussed earlier
in this section.
In the frequency region A in Fig. 4, the beat frequency due
to an interharmonic is not properly represented by the IEC
ickermeter. Note that the local minima point in the A region is
at 4.4 Hz, which is half of the maximum response frequency
(8.8 Hz) of the weighting lter in Block 3. Furthermore, as
discussed earlier, only the squared value of the relative inter-
harmonic magnitude appears in this frequency region A, so that
the inuence of the interharmonics in this region is signicantly
reduced.
The IEC ickermeter seems to properly work for inter-
harmonics in region B, but as the interharmonic frequency
becomes greater than 102 Hz in region C, the interharmonic
magnitude required to generate perceptible icker rapidly
grows, and it goes beyond the 10% of the fundamental fre-
quency magnitude after about 110 Hz. However, according to
the real eld case reported in [8], light icker was observed by
residential customers when 180- to 200-Hz interharmonics with
1% magnitude of the fundamental component were present,
thereby conrming the deciency of the IEC ickermeter stan-
dard in the high-frequency range.
Based on our analysis and numerical experiments, the de-
ciency of the IEC ickermeter with regard to icker caused by
interharmonics has been demonstrated. To address this prob-
lem, we propose a new approach based on down-up sampling
of the input voltage signal in the next section.
IV. DOWN-UP SAMPLING-BASED
PEAK DETECTION METHOD
With interharmonics present, there are two types of uctu-
ations due to the interharmonics (RMS and peak value uc-
tuations) [6]. According to the experimental study reported
in [4], the icker behavior widely varies, depending on the
type of lamp. Analysis results in [6] suggest that certain types
of lamps (e.g., compact uorescent lamps) are more sensitive
to peak value uctuations of the input voltage rather than
RMS value uctuations. Furthermore, even small magnitude
interharmonics that do not cause signicant RMS uctuations
can result in considerable peak uctuations. Therefore, for a
given interharmonic amplitude, a possible worst case might
occur if the customers lighting system consists mainly of peak-
uctuation-sensitive lamps.
Hence, in this paper, we propose alternative signal processing
blocks to detect peak uctuations for icker evaluation. Fig. 5
presents the signal processing blocks for our proposed icker-
detection method. As we have already mentioned, light icker
due to peak value uctuations is caused by modulation imposed
on the absolute value of the voltage signal, which has the
second harmonic of the fundamental frequency as the dominant
frequency term. Therefore, our basic idea is to examine the
peak uctuation imposed only on the second harmonic of the
fundamental frequency.
First, the input voltage signal is sampled by the sampling
frequency f
S
= Mf
1
, as schematically indicated in Block A in
Fig. 5. The sampling frequency should be an integer multiple M
of the fundamental frequency to prevent spectral leakage
around the fundamental term. The IEC standard allows either
analog or digital implementation of the recommended icker-
meter [2], and ickermeters are digitally implemented using
various approaches in the literature, including [14][17].
Then, the absolute values of the samples are generated, as
shown in Block B in Fig. 5. Assuming that the voltage signals
are processed in the discrete-time sample domain, we can assess
the peak uctuations of the absolute-valued voltage signals by
downsampling to 2f
1
or downsampling by (2/M), as shown
in Block C in Fig. 5. The downsampling process actually
corresponds to taking one sample for each half-cycle of the
fundamental. The process can be represented in an analytical
way using the signal model in (1).
First, the analytical expression of the absolute value of
the voltage signal should be considered. We can consider the
absolute values of a sinusoidal signal as the multiplication
between the sinusoidal signal and a square-wave function with
the same period. Assuming that we have a sinusoidal signal
sin(2f
1
t), then the Fourier series representation of the cor-
responding square-wave function s(t) is as follows:
s(t) =
4

n=1,3,...
1
n
sin(2nf
1
t) (4)
where n is an odd integer. Therefore, the absolute values of the
signal model in (1) are given in (5), shown at the bottom of
the next page, where the v
f
(t) part indicates the terms related
to the fundamental frequency, and the v
IH
(t) part indicates
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE UBERLANDIA. Downloaded on May 11,2010 at 00:31:57 UTC from IEEE Xplore. Restrictions apply.
KIM et al.: DETECTION OF FLICKER CAUSED BY INTERHARMONICS 157
the terms related to the interharmonics. Then, the discrete-time
version of the v
f
(t) part can be represented as follows:
v
f
[k] =v
f
_
t
0
+
k
2f
1
_
=
2

n=1,3,...
1
n
_
cos
_
2(n 1)f
1
_
t
0
+
k
2f
1
__
cos
_
2(n + 1)f
1
_
t
0
+
k
2f
1
___
=
2

n=1,3,...
1
n
_
cos (2(n 1)f
1
t
0
+ (n 1)k)
cos (2(n+1)f
1
t
0
+(n+1)k)
_
(6)
where k denotes the discrete-time index, and t
0
is the initial
sampling time. v
f
[k] becomes a dc term (a constant value)
since the 2(n 1)f
1
t
0
terms in the cosine functions are
independent of k, and the other terms in the cosine functions,
i.e., (n 1)k, become 2lk(l = 0, 1, 2, . . .) due to the fact
that n is an odd integer.
According to (2), f
IH
can be represented as hf
1
f
B
(h =
2p + 1; p = 0, 1, 2, . . .), and then, the discrete-time version of
v
IH
in (6) becomes the following:
v
IH
[k] =v
IH
_
t
0
+
k
2f
1
_
=
2m

n=1,3,...
1
n
cos
_
2(hf
1
f
B
nf
1
)

_
t
0
+
k
2f
1
_
+
IH
_

2m

n=1,3,...
1
n
cos
_
2(hf
1
f
B
+ nf
1
)

_
t
0
+
k
2f
1
_
+
IH
_
=
2m

n=1,3,...
1
n
_
cos
_
2l
f
B
k +
n

IH
_
cos
_
2l
f
B
k +
n
+
IH
_
_
(7)
where
n

IH
= 2((h n)f
1
f
B
)t
0
+ (h n)k +
IH
, and

n
+
IH
= 2((h + n)f
1
f
B
)t
0
+ (h + n)k +
IH
. Note that
l
f
B
(= (f
B
/2f
1
)) corresponds to the discrete frequency rep-
resentation of the actual beat frequency in (2). Hence, v
IH
[k]
represents the magnitude m of the actual beat term and its
frequency f
B
.
Consequently, we can nd a frequency term that represents
the actual beating effect via the downsampling process. How-
ever, there are potential problems if only the downsampling
method is used. For example, the remaining parts of the ick-
ermeter should be changed for the downsampled signal. More-
over, to obtain the proper accuracy recommended by the IEC
specications, the sampling frequency should be sufciently
high [14], [16]. Additionally, we want to replace those signal
processing blocks in the IEC standard, which potentially cause
problems in detecting icker caused by interharmonics and
seamlessly combine the newly proposed blocks with the rest
of the ickermeter blocks, which comply with the current IEC
standard.
Therefore, we up-sample the downsampled signal to the
original sampling frequency, as indicated in Block D in Fig. 5.
The upsampling process is done by inserting zeros between
samples, which is known as zero padding [18]. This down-
up sampled signal carries the peak uctuation information as
sideband terms around each harmonic of the fundamental term.
The following Block E in Fig. 5 is a bandpass lter that
removes the dc term and unnecessary frequency terms resulting
from the upsampling. In this paper, we used a rst-order high-
pass lter whose cutoff frequency is 0.05 Hz and a sixth-order,
low-pass lter whose cutoff frequency is 60 Hz for 60-Hz
systems and 50 Hz for 50-Hz systems, which are half of the
downsampling frequency. Note that the high-frequency cutoff
of this bandpass lter is different from that of the bandpass lter
in the IEC ickermeter.
Blocks AE in Fig. 5 comprise the proposed alternative sig-
nal processing blocks. The proposed signal processing blocks
replace Block 2 and the rst bandpass lter in Block 3 of the
IEC standard ickermeter in Fig. 1. The rest of the proposed
ickermeter, including the weighting lter in Fig. 5, is the same
as the IEC standard ickermeter.
|v(t)| =v(t) s(t)
= (sin(2f
1
t) + msin(2f
IH
t +
IH
))
_
4

n=1,3,...
1
n
sin(2nf
1
t)
_
=
4

n=1,3,...
1
n
sin(2f
1
t) sin(2nf
1
t) +
4

n=1,3,...
m
n
sin(2f
IH
t +
IH
) sin(2nf
1
t)
=
4

n=1,3,...
_
1
2n
cos (2(n 1)f
1
t)
1
2n
cos (2(n + 1)f
1
t)
_
. .
v
f
(t)
+
4

n=1,3,...
_
m
2n
cos (2(f
IH
nf
1
)t +
IH
)
m
2n
cos (2(f
IH
+ nf
1
)t +
IH
)
_
. .
v
IH
(t)
(5)
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE UBERLANDIA. Downloaded on May 11,2010 at 00:31:57 UTC from IEEE Xplore. Restrictions apply.
158 IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, VOL. 58, NO. 1, JANUARY 2009
Fig. 6. Down-up sampling-based icker detection with an 8-Hz interharmonic term. (a) Discrete time-domain signal after Block D in Fig. 5. (b) Corresponding
DFT magnitude spectrum.
Note that our method can be considered conservative in
assessing the likelihood of icker, particularly when it is used
for assessing icker of lighting systems, which consist mainly
of incandescent lamps. That is because our method evaluates
the peak uctuations of the voltage input, while incandescent
lamps are sensitive to the RMS value uctuations. As discussed
in [6], peak values generally tend to uctuate more severely
than RMS values do in the presence of the same amplitude
interharmonic.
We implemented our new proposed algorithm using
MATLAB, and the experimental results based on the implemen-
tation are presented in the next section.
V. EXPERIMENTAL RESULTS
To demonstrate the efcacy of the proposed method, we
rst show a proper icker-detection example using the same
signal model in Section III, i.e., the 60-Hz voltage signal with
an 8-Hz interharmonic. The signal is sampled at 1200 Hz
(= 60 20), which also prevents spectral leakage around the
fundamental term. Fig. 6 illustrates the signal processing result
of the proposed alternative blocks. Fig. 6(a) shows the discrete-
time waveform after Block D in Fig. 5, which is the down-
up sampled signal, and Fig. 6(b) presents the corresponding
frequency spectrum result.
Contrary to the misleading frequency spectrum result
in Fig. 2(b), which showed a 16-Hz term instead of
52(= 60 8) Hz, we can successfully detect the 52-Hz uc-
tuation term in the DFT magnitude spectrum of Fig. 6(b). The
other higher frequency terms and the dc term will be ltered out
by a bandpass lter in Block E in Fig. 5. The 52-Hz component
is then processed by statistical processing units to generate
icker-severity indexes.
An example of detecting icker caused by a high-
frequency interharmonic (172 Hz) is also presented in Fig. 7.
As mentioned earlier, the current IEC standard ickerme-
ter cannot detect the icker caused by this high-frequency
interharmonic since the higher frequency term is ltered out
by a bandpass lter in Block 3 of the IEC ickermeter in Fig. 1.
However, the down-up sampled signal presented in Fig. 7(a)
and the corresponding DFT magnitude in Fig. 7(b) accurately
show the actual beating 8(= 3 60 172) Hz term.
For a quantitative comparison, the sensitivity curve of the
proposed method to interharmonics is presented in Fig. 8, along
with that of the IEC ickermeter. These curves represent the
minimum interharmonic magnitude m
min
to produce one PU in
the ickermeter output. The sensitivity curves for both methods
appear identical in the frequency range from around 30 to 90 Hz
in Fig. 4, where the IEC ickermeter properly works with
interharmonics.
However, the sensitivity curves for both methods exhibit
different behaviors in the lower frequency range below 30 Hz
and the higher frequency range above 90 Hz. Contrary to the in-
accurate response of the IEC ickermeter in the low-frequency
range (018 Hz), as discussed earlier, the curve corresponding
to the newly proposed method shows that the proposed method
sensitively reacts to the low-frequency interharmonics in that a
smaller interharmonic magnitude is required to produce one PU.
In the high-frequency range (90120 Hz), according to the
curve corresponding to the standard method, large magnitudes
of the interharmonic are required to produce one PU in the
ickermeter output, which is contrary to the eld case reported
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE UBERLANDIA. Downloaded on May 11,2010 at 00:31:57 UTC from IEEE Xplore. Restrictions apply.
KIM et al.: DETECTION OF FLICKER CAUSED BY INTERHARMONICS 159
Fig. 7. Down-up sampling-based icker detection with a 172-Hz interharmonic term. (a) Discrete time-domain signal after Block D in Fig. 5. (b) Corresponding
DFT magnitude spectrum.
Fig. 8. Minimum interharmonic magnitude m
min
necessary to produce
one PU for the IEC ickermeter response and the proposed method versus
interharmonic frequencies.
in [8]. On the other hand, the curve corresponding to the pro-
posed method is well extended into the high-frequency range
and shows that relatively small magnitudes of the interharmonic
can produce one PU.
The numerical experimental results presented in this section
indicate that our proposed icker processing blocks are able to
properly detect ickers caused by interharmonics, regardless of
their frequencies.
VI. CONCLUSION
In this paper, we have presented a deciency of the current
IEC ickermeter with regard to low-frequency interharmonics
and propose a new approach based on down-up sampling of
the voltage uctuation signal to address this deciency. Using
the down-up sampling-based method, peak value uctuations
of the input voltage were examined to assess interharmonic-
caused icker.
Based on our analysis and numerical experiments, it was
demonstrated that our proposed method correctly represents
the actual beat frequency associated with the interharmonics
in any frequency range. As a result, the proposed method can
extend the interharmonic-related icker-detection range beyond
the current limited range of the IEC ickermeter.
Since our method is based on evaluating peak value uctu-
ations, it is particularly suitable for assessing icker of peak-
uctuation-sensitive lamps, e.g., compact uorescent lamps. In
this sense, our method may be considered to be complementary
to the IEC standard that is focused on assessing icker of incan-
descent lamps, which are sensitive to RMS value uctuations.
Furthermore, our proposed signal processing blocks can be
immediately combined with the current ickermeter standard
to address the interharmonic-related limitations.
ACKNOWLEDGMENT
The authors would like to thank the anonymous reviewers for
their constructive suggestions.
REFERENCES
[1] S. Clippingdale and H. Isono, Photosensitivity, broadcast guidelines and
video monitoring, in Proc. IEEE Int. Conf. Syst., Man, Cybern., Tokyo,
Japan, Oct. 1999, pp. 2227.
[2] Electromagnetic Compatibility (EMC). Part 4: Testing and measurement
techniques. Section 15: Flickermeter. Functional and Design Specica-
tions, IEC Std. 61 000_4_15, 2003.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE UBERLANDIA. Downloaded on May 11,2010 at 00:31:57 UTC from IEEE Xplore. Restrictions apply.
160 IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, VOL. 58, NO. 1, JANUARY 2009
[3] IEEE Recommended Practice for Measurement and Limits of Voltage
Fluctuations and Associated Light Flicker on AC Power Systems, IEEE
Std. 1453_2004, 2004.
[4] M. D. Koster, E. D. Jaiger, and W. Vancoistem, Light Flicker Caused
by Interharmonics. [Online]. Available: http://grouper.ieee.org/groups/
harmonics/iharm/docs/ihicker.pdf
[5] T. Keppler, N. R. Watson, J. Arrillaga, and C. Shiun, Theoretical assess-
ment of light icker caused by sub- and interharmonic frequencies, IEEE
Trans. Power Del., vol. 18, no. 1, pp. 329333, Jan. 2003.
[6] T. Tayjasanant, W. Wang, and W. Xu, Interharmonicicker curves,
IEEE Trans. Power Del., vol. 20, no. 2, pp. 10171024, Apr. 2005.
[7] W. Xu, Deciency of the IEC icker meter for measuring interharmonic-
caused voltage ickers, in Proc. IEEE Power Eng. Soc. Gen. Meeting,
Jun. 2005, pp. 22852288.
[8] T. Tayjasanant and W. Xu, A case study of icker/interharmonic prob-
lems caused by a variable frequency drive, in Proc. 11th Int. Conf.
Harmonics Quality Power, Sep. 2004, pp. 7276.
[9] D. Gallo, R. Langella, and A. Testa, Light icker prediction based
on voltage spectral analysis, in Proc. IEEE Porto Power Tech Conf.,
Sep. 2001, pp. 1013.
[10] D. Gallo, R. Langella, and A. Testa, Toward a new ickermeter based
on voltage spectral analysis, in Proc. IEEE Int. Symp. Ind. Electron.,
Jul. 2002, pp. 573578.
[11] E. W. Gunther, Interharmonics in power systems, in Proc. IEEE Power
Eng. Soc. Summer Meeting, Jul. 2001, vol. 2, pp. 813817.
[12] F. D. Rosa, R. Langella, A. Sollazzo, and A. Testa, On the interharmonic
components generated by adjustable speed drives, in Proc. IEEE 10th
Int. Conf. Harmonics Quality Power, Oct. 2002, vol. 1, pp. 183188.
[13] T. Kim, A. Wang, E. J. Powers, W. M. Grady, and A. Arapostathis,
Detection of icker caused by high-frequency interharmonics, in Proc.
IEEE Instrum. Meas. Technol. Conf., Warsaw, Poland, May 2007, pp. 15.
[14] J. de Sousa Rocha and S. Deckmann, Digital ickermeter implemen-
tation, in Proc. 38th Midwest Symp. Circuits Syst., Aug. 1995, vol. 2,
pp. 757760.
[15] M. Chen, Digital algorithms for measurement of voltage icker, Proc.
Inst. Elect. Eng.Generation, Transm. Distrib., vol. 144, no. 2, pp. 175
180, Mar. 1997.
[16] S. Caldara, S. Nuccio, and C. Spataro, A virtual instrument for measure-
ment of icker, IEEE Trans. Instrum. Meas., vol. 47, no. 5, pp. 1155
1158, Oct. 1998.
[17] A. Bertola, G. Lazaroiu, M. Roscia, and D. Zaninelli, AMatlabSimulink
ickermeter model for power quality studies, in Proc. 11th IEEE Int.
Conf. Harmonics Quality Power, Sep. 2004, pp. 734738.
[18] A. V. Oppenheim and R. W. Schafer, Discrete-Time Signal Processing.
Englewood Cliffs, NJ: Prentice-Hall, 1999.
Taekhyun Kim(S03) received the B.S. degree from
Seoul National University, Seoul, Korea, and the
M.S. degree from The University of Texas at Austin,
where he is currently working toward the Ph.D.
degree in electrical and computer engineering.
His research interests include the theory and al-
gorithm development of advanced signal processing
techniques for power system analysis.
Edward J. Powers (F83) received the B.S. degree
from Tufts University, Medford, MA, the M.S. de-
gree from the Massachusetts Institute of Technology,
Cambridge, and the Ph.D. degree from Stanford Uni-
versity, Stanford, CA, all in electrical engineering.
He is the Texas Atomic Energy Research Foun-
dation Professor of Engineering and a Professor of
electrical and computer engineering with The Uni-
versity of Texas at Austin. His primary professional
interests are in the innovative application of digital
higher order statistical signal processing in the analy-
sis, interpretation, and modeling of time series data characterizing nonlinear
physical phenomena, e.g., interharmonics and health monitoring of rotating
machines, and the utilization of the wavelet transform and timefrequency
analysis to detect and identify transient events in various physical systems,
including electric power systems.
W. Mack Grady (F00) received the Ph.D. degree
from Purdue University, West Lafayette, IN.
He is a Professor of electrical and computer engi-
neering with The University of Texas at Austin. He
worked for six years as a System Planning Engineer
for Texas Utilities, Dallas.
Prof. Grady is a Chairman of the IEEE Power
Engineering Society (PES) Transmission and Distri-
bution Committee and was the Technical Program
Chairman for the 2003 IEEE PES Transmission and
Distribution Conference and Exposition.
Ari Arapostathis (F07) received the B.S. de-
gree from Massachusetts Institute of Technology,
Cambridge, and the Ph.D. degree fromthe University
of California, Berkeley.
He has been a Faculty Member with The Uni-
versity of Texas at Austin since 1982. His research
interests include analysis and estimation techniques
for stochastic systems, the application of differential
geometric methods to the design and analysis of
control systems, stability properties of large-scale
interconnected power systems, and stochastic and
adaptive control theory.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE UBERLANDIA. Downloaded on May 11,2010 at 00:31:57 UTC from IEEE Xplore. Restrictions apply.